25th Umbrella Symposium For the development of Joint Cooperation Ideas “Modeling and Simulation with emphasis on high performance computing and grid computing” Visualization of Probabilistic Fiber Tracts in Virtual Reality

Understanding the connectivity structure of the human brain is a fundamental prerequisite for the treatment of psychiatric or neurological diseases. Probabilistic tractography has become an established method to account for the inherent uncertainties of the actual course of fiber bundles in magnetic resonance imaging data. This work presents a visualization system that addresses data set sizes that fit into graphics memory (up do 1.5GB), currently the system supports: (1) The visualization emphasizes spatial patterns and present the three-dimensional physical structure in an intuitive fashion via a virtual reality interface. (2) The real-time renderings convey the uncertainty within each fiber tract by using color and opacity modulation of direct volume rendering techniques. (3) The location of fiber tracts within the human brain can easily be deduced by the anatomical context in form of common two-dimensional slices as well as brain areas. (4) The understanding of the structure-function relationship is supported by interactive manipulation techniques based on magic lens interaction metaphors. (5) All of the above is integrated in an interactive exploration system. Future challenges will result from significantly increased system requirements. Next generation data sets are expected to greatly exceed the 20 TB threshold. Single workstation will no longer be sufficient to interactively process, less visualize such data sets. In particular, the concurrent visualization of multiple data modalities, that include but are not limited to cytoarchitecture, MR, PET, brain areas and fiber tracts (both DTI and PLI), will be a crucial step to help understanding complex and interconnecting phenomena and will have to be addressed in the future.